Butterflies as indicators: responses of butterfly populations to climatic fluctuations in Germany Martin Wiemers, Alexander Harpke, Elisabeth Kühn, Martin Musche, Oliver Schweiger, Josef Settele, Marten Winter

Background

• Research project on monitoring climate change and biodiversity in the German Free State of Saxony

• Development of biodiversity indicators for climate change

• Evaluation of available data • Impact of climate change on

• Community composition • Distribution ranges

SEITE 2

Temperature trend in Dresden 1812-2009

SEITE 3

Responses of species communities to climate change

• Community Temperature Index (CTI) - Devictor et al. (2008)

• Weighted Sum of Species Temperature Indices (STI)

SEITE 4

CTI

Community Temperature Index in Butterflies

• Positive temporal trend of CTI in European Butterflies 1990-2009 (Van Swaay et al. 2008, 2010)

• European variations in trend of bird & butterfly CTI (Devictor et al. 2012 in Nature Climate Change)

SEITE 5

SEITE 6

Annual mean temperatures in Europe: Basis for Species Temperature Indices (STI)

Reference period: 1961-1990 Leipzig: 8.8 Sachsen: 8.1

SEITE 7

STI examples in butterflies

Colias palaeno: 3.62

Hipparchia statilinus: 11.82

Boreo-alpine species:

Mediterranean species:

Butterflies of Saxony • 140 species • Good temporal coverage for 36 years

(1975-2010) with >1000 records/year • Excellent spatial coverage • 43% of records without abundance data

SEITE 8

Butterfly Monitoring Germany (TMD) • 149 species • Monitoring data • Temporal coverage for 5 years

(2006-2010) • Limited spatial coverage

SEITE 10

CTI correlation with climatic fluctuations

SEITE 11

Correlation with annual mean temperature: 0.55, p<0.001***

▬ Saxony annual mean temperature ▬ CTI

Considering generation time

SEITE 13

Better Correlation (0,64) with the sliding 3-year mean of the current and 2 previous years

Generation length up to one year

Temperature of previous years decisive

▬ Annual mean temperature ▬ Mean temp. of the last 3 years ▬ CTI

Year x 0.549 *** 0.632 ***

0.637 *** 0.584 ** x-1 0.496 ** 0.491 **

0.424 * x-2 0.324 0.260 x-3

Correlation between CTI and annual mean temperatures across Europe

SEITE 14

0 5 10 15

05

1015

Mean annual temperature

CTI

Nearly linear correlation with an increase of 0.33

Saxony 1975-2010: Increase of

ann. mean: 0.0288 Projected CTI: 0.0095 Observed CTI: 0.0046 Difference: factor 0.5

Robustness of CTI

SEITE 15

CTI is robust to selection of species Exceptional immigration events may influence CTI

▬ without Colias crocea ▬ with Colias crocea

CTI – TMD Regions

SEITE 18

Yearx -0.169x-1 0.940 **x-2 0.518x-3 -0.316

0.558

0.1300.944 **

0.7180.825 *

0.781

• Strong correlation between CTI and the temperature mean of the previous year

SEITE 19

Climate impact on regional distribution ranges

Saxony: Southern species in the northern flatlands Northern species in the mountains near the

southern border to the Czech Republic

Hipparchia statilinus: warm-adapted species

Colias palaeno: cold-adapted species

STI: 11.8 STI: 3.6

SEITE 20

STI-based Areal Index Sum of occupied grid cells (e.g. MTB) of

southern versus northern species based on and weighted by the difference of each species‘ STI from the mean CTI

Examples of STI differences: The butterfly genus Colias: Colias alfacariensis 9.94 1.176Colias crocea 10.69 1.923Colias erate 10.03 1.266Colias hyale 8.37 -0.397Colias palaeno 3.62 -5.146

Butterflies of Saxony: STI-based Areal Index

SEITE 21

Correlation of areal index with annual mean temperature (0,48**) Positive trend

Areal Index vs. CTI in Butterflies of Saxony

SEITE 22

Very high correlation AISTI ↔ CTI: 0,93***

▬ Areal Index (STI-based) ▬ CTI

Conclusions • CTI indicates change of community composition

towards warm-adapted species during the last 2-3 decades

• CTI is a robust biodiversity indicator for climate change, but works best with monitoring data

• CTI is sensitive to short-term climatic fluctuations in animals with short generation times

• The extent of CTI fluctuations strongly depends on generation length

• STI values can be used to assess climate-driven range changes

• Climate-driven range changes strongly correlate with changes in community composition

SEITE 23

Thank you for your attention!

• Recorders and contributors to the Butterfly Fauna of Saxony and the Butterfly Monitoring Scheme Germany (TMD)

• Rolf Reinhardt and the „Sächsisches Landesamt für Umwelt, Landwirtschaft und Geologie (LfULG)“ for data

• Ingolf Kühn and members of the Department of Community Ecology for helpful discussions

SEITE 25

Acknowledgements